This proposal is aimed at a systems-level analysis of how the gastric mill central pattern generator (CPG) of lobsters functions as a neuronal network and how it integrates sensory feedback and coordinating central inputs in a complex, plastic manner. While not immediately related to health sciences, the gastric CPG nevertheless is an excellent system for studying basic mechanisms underlying the generation and control of rhythmic behaviors. Since 1969 this work has established the lobster stomatogastric ganglion as the most thoroughly understood neural circuit presently available. The research plan has 3 specific aims: 1. To study the role of synaptic timecourse in the temporal operation of the gastric CPG. Electrophysiological and pharmacological methods will be used to study the transmitters and postsynaptic (possibly metabotropic) mechanisms underlying a recently-discovered class of slow synapses. Pharmacological manipulation will be used to dissect the role of synaptic timecourse in temporal processing within this neural circuit. 3. To examine plasticity of sensorimotor processing in the gastric CPG. Here recent studies of identified mechanosensory feedback loops will be extended. Peptidergic and muscarinic modulators are known to produce a functional reorganization of the gastric circuit. Using electrophysiological techniques to record and manipulate activity unidentified neurons, and a dye-sensitized photinactivation method to ablate specific members of the network, this project will analyze the cellular and synaptic basis of changes in sensorimotor processing within the circuit, induced by neuromodulators. 3. To study the actions on the gastic CPG of a pair of identified, command interneurons that serve to coordinate its rhythmic operation with that of an adjacent motor system. Current-and voltage-clamp studies will be combined with pharmacological techniques to extend the understanding of the transmitters and membrane responses involved in the multiple synaptic actions of these interneurons. These cellular and synaptic properties will be related to a systems-level analysis of how the coordinating inputs alter the function of the whole network.